topbanner.gif
Login
Username

Password




Not a member yet?
Click here to register.

Forgotten your password?
Request a new one here.
Shoutbox
You must login to post a message.

JackH
OfflineAdmin
· 07/04/2020 10:27 AM
Happy 4th of July! jazzyhands

JackH
OfflineAdmin
· 06/24/2020 7:58 AM
@Mark McCornack, Please post your question in the forum.

Mark McCornack
Offline
· 06/15/2020 9:28 PM
Hi! Looking for a legacy inlet temp sensor on 13 yr old Gene Cafe. It seems they've changed it and now you need new mother board and new sensor. Any ideas where I can find compatibile old one? Mark

Samaniego
Offline
· 06/09/2020 6:39 PM
Wich thermometers Can i buy for my roasting machine compatible with usb or macbook?

JackH
OfflineAdmin
· 06/05/2020 5:38 PM
peveleth, It is better if you start a post in the forum with your question. These shouts go away in time.

Users Online
Guests Online: 5

Members Online: 0

Total Members: 6,636
Newest Member: CyrusF
In Memory Of Ginny
Donations

Latest Donations
Anonymous - 5.00
Anonymous - 5.00
renatoa - 2.00
JitterzZ - 2.01
renatoa - 2.00

View Thread

Who is here? 1 guest(s)
 Print Thread
TC4C with J Type Floating Data
ged
Hi everyone. I've been using Roast Logger for a few years and love it to bits, (my eternal gratitude to all those responsible!) but I changed roasters almost a year ago and haven't been able to get my new setup running, so I'm hoping to get some help.

The new(er) roaster is a Diedrich IR7 which runs J type thermocouples and displays in Fahrenheit. I'm doubling my thermocouple connection from the Diedrich thermocouple where it connects into its controller, then using shielded J type wire into the TC4C. The shield is attached to the Diedrich shield at the roaster end and not connected at the TC4 end.

The roaster presently has only a single thermocouple so I have a second J type thermocouple reading ambient temp to provide a second data stream.

I've altered the user.h file in the TC4 sketch to read J types and I'm running a jumper on the TC4 Anlg2 pins to provide Fahrenheit outputs.'

At start-up the roaster temp display and the Roast Logger temp displays are about equal but then diverge by about 10 degrees per load until they vary by about 70 degrees eventually.

I've suspected a ground loop somewhere that might be causing the issue, but I've run various configurations of the shielding etc. and nothing seems to solve it.

I've noticed a couple of other threads here with similar symptoms, but haven't seen a definitive solution.

So anyway, after buggarising around with it for almost a year, I'm desperate to get it running again! Any suggestions would be greatly appreciated.
 
BenKeith
One suggestion, if you are using shielded thermocouples, I would suggest not grounding both ends of the shield. This can sometimes establish a current path through the shield that can cause a number of problems.

What version RoastLogger are you running in the computer and RoastloggerTC4 are you running in the TC4.

I've also been running RL for several years and have never had a problem until I recently installed the RL Ver 2.4 and the Latest RLTC4. Last week I did two roast and my fan kept cutting back to 86%. You could tell it was the PID causing it because the slider stayed on 100% but the LCD showed it was 86%. I could move the slide back to about 50% and back up to 100% and the fan would go back to 100% for several seconds and then cut back to 86%.
The other day, I was using the same laptop but a different TC4 with the Latest version of RL-TC4 loaded in it, Again, two roast and it did the same thing I ended up unplugging the fan from the relay and plugging it directly into a drop cord. I'm going to download fresh versions of both Sketches and try them again. If I get the same results, I'm going back to the earlier version I was using.

Wait now, I just read your post again and caught you saying you are running the jumper. DON'T do that, remove that jumper. Go into use.h and right at the beginning where it has the F or C choice, Comment out the C selection. Then go into the compture and select to use F in there. That should give you F. readings

One other thing, the reason I never give temps of roast, there are way to many variables. You probe place being just a few inches from each other than cause huge differences. The probes themselves the have big differences. If the shift is the same, that adjust for it. I'm using a probe right now that reads 15 degrees lower than my other probe did in the exact same spot than my other probe did, First couple of roast I used it went into second crack heavily at 330 degrees, but didn't even hit SC until 442 or so with the other one.
Edited by BenKeith on 01/27/2018 6:44 PM
 
ged
Yeah, it has all the hallmarks of a ground loop doesn't it.

I've tried a couple of configurations but presently have it unattached at the TC4 end and tied to the shield of the Diedrich thermocouple at the roaster end. The Diedrich shield finishes about 75mm short of the control board, so I've done the same and then taped the shields together to ground them. I've run them unattached as well.
 
ged
Oh, ok. I've just seen the last paragraph of your post. Interesting!

I think I'm running the same original version of Roast Logger that I've used for 3 or 4 years. Just waiting for the old laptop to start up so I can check....

I've never run any of the PID functions or controls, so that side of it is new to me. If I can ever get it logging correctly! I intend using the PID functions.

Ill check my versions.
 
BenKeith
I've added so much to my first reply, I figured I would start another.
One thing I didn't think about since I run the pots on all mine. If you don't run the pots, it's recommend you put 10K pull down resistors on ANL-1 and ANL-2.

OK, I've never used it as a Data Logger. When I first started using it, it was for the PID control. I've gotten so used that using the PID, I don't ever try to roast without it anymore. LUV that thing.
Edited by BenKeith on 01/27/2018 6:55 PM
 
ged
Im running,

Roast Logger 2.3.5

Roastlogger TC4 Arduino 1.0.5-R2
 
BenKeith
The RoastLogger ain't bad, he hasn't make tons of changes to it. The RLTC-4 have had tons of changes to it. Ver 3.4 is the latest one, 3.3 was the one I was using and might be going back to.
Understand also, there were some significant changes to Arduino a couple of years ago, so if you update, you have to update all the libraries also because the older versions don't work with Arduino when it gets past about 6.13 or something like that. Also, the newer version won't work with the older versions of Arduino.
 
ged

Quote

BenKeith wrote:

I've added so much to my first reply, I figured I would start another.
One thing I didn't think about since I run the pots on all mine. If you don't run the pots, it's recommend you put 10K pull down resistors on ANL-1 and ANL-2.

OK, I've never used it as a Data Logger. When I first started using it, it was for the PID control. I've gotten so used that using the PID, I don't ever try to roast without it anymore. LUV that thing.


Haha. I wondered why I hadn't seen that part!

Ok, so your saying that the PID controls run off Anlg1 and Anlg2 and you run potentiometers on those connections to control fan settings etc? and I should run either a pot or a pull down resistor on them?

I'm running through your other suggestions there too Ben. Ill be back! Thanks.
 
BenKeith
Pots are not required, you can use the sliders built into the RL program. I use the pots sometimes if I want to make a quick change. However doing that, when the PID hits it's next programed change, it's going to take control back again, and if you want to override it with the pots, that's doable, until it hits it's next change, so, using the pots, your are at war with the PID and it's usually going to win.
I also have one TC4 set up for a stand alone unit that I don't hook to the computer, so the pots and buttons get used on that one.

As far as PID control, that's in how you set it up. Once you get it right, you don't do anything but start and stop it (eject the beans), it's does everything between those to points
 
ged
Ok, so I’ve updated the Arduino ide to the latest version, updated the TC4 sketch to the most recent version (with the AC fan control). I’ve not updated the Roast Logger version though. I think I’m already using the latest version.

Same same unfortunately. Readings were about 1 degree different at startup, then drifted to be 8 degrees different at load then about 20 degrees apart by the end of the 14 minute roast cycle. I didn’t keep running it beyond that, but I think as time increases the disparity increases. It’s got to be some sort of compounding loop doesn’t it?
 
BenKeith
After reading your post a little closer, the first time I read too much into it. I thought you were using two thermocouples, not paralleling two set of electronics of one probe. I've never tried that, so can't comment much about it.

I can see where there would be all kinds of potential for problems. Unless you were able to get the exact same wire for both connections, that alone could cause problems. I once tried to switch two thermocouples with a toggle switch. The monitor I had would only read four and I was needing to read six on an outboard motor. That didn't work, just having the switch in line screwed all the readings up.

I would have opted for coming up with some way to get another thermocouple in there, don't think I would have ever tried to parallel of the one in it.
Maybe one of the other drum roaster members can give you some suggestions. You are doing something I've never done and wouldn't have never tried doing because from my experience with them has never been good when doing what I shouldn't.
 
ged
Yep. The parallel issue could be at its heart and it’s a good point regarding different resistances in the cabling. I stole that idea off a YouTube vid where a guy runs a parallel connection from an IR1 into Typica via a controller of some sort.

Given that I’ve eliminated (nearly) as many potentials as I can in my setup, I’m starting to suspect the board... I’m pretty sure this was shorted across some of its back pins a couple of months ago and I’m not entirely confident it is still performing correctly. I don’t think there is a test for the TC4 to establish if it’s functional and accurate?

I’ve emailed Jim this morning to see if he has any boards available at the moment. I could use a spare either way. Failing that, I’m looking around at switching to another platform entirely. Typica and Phidgets looks doable. But it’s a shame. I like Roast Logger, but I need a solution.

Thanks for your help BenKeith.
 
BenKeith
You can get a couple of those cheap Chinese thermocouples off ebay for c couple of dollars. Water boils at 212 degrees at sea level so that gives you a fairly reliable test temp. Try it in some boiling water and see how close it is there.
 
greencardigan
I'd try the boiling water like Ben suggested. Try with each device connected separately then with both in parallel. See if you get different readings on the TC4 when separate compared to in parallel.
 
Will2

Quote

BenKeith wrote:
...Unless you were able to get the exact same wire for both connections, that alone could cause problems. ...

Yes.
Tested?
Try to replace the wires between themselves (at both ends).
Do not rely on the color label.
Is really used j-type wire?
Can you take pictures?
Edited by Will2 on 01/30/2018 6:06 AM
Viliam
 
ged
Thanks for the replies guys. I’ve been a bit crook for a couple of days, thus the delayed responses.

Ok, I’m battling a little with the logic of Ben’s suggestion re probes in boiling water. I think I’m not explaining the set up properly....

There is only a single (BT) thermocouple, which is the Diedrich standard arrangement. I was running a second thermocouple reading ambient temperature, just to provide a data flow to the sketch, but for the sake of clarity, I’ve disconnected that TC and set Roastlogger to disable charting of T2.

So, the single thermocouple runs to the Diedrich instrument amplifier and LED temp display on the side of the machine. I’m taking my thermocouple feed from the connections at this point and feeding that to the TC4 via 1m of white/red, insulated J Type thermocouple wire. Yes, it is specific J Type wire which I have acquired specifically for this purpose.

I’ve copied this arrangement from this YouTube video:

https://youtu.be/EXPjLgCoUWU

You’ll notice he is using a National Instruments amplifier/interface. Whether that runs differently to the TC4, I’m not too sure, but I would have said electrically, they are the same arrangement except for the added resistance of the extra 1M of cabling, which given the tiny voltages being measured, maybe significant. I’m thinking that may account for small differences in the displayed temps at startup between the roaster display and the Roastlogger display, but the compounding error?

I still suspect the culprit will be the TC4 board... Ive no way of verifying it’s accuracy. I have inadvertently killed another TC4 by leaving it connected to the computer for a few days and then getting a big storm and lightening strikes, and it never ran properly after that. Ive retested that board as well in this setup and get no data.

I’ll try and get some photos today which might make things a bit clearer. Thanks for the assistance.
 
BenKeith
The boiling water suggestion was to get a couple of those cheap ebay thermocouples to test the TC-4. I realize you are not going to take the one out of the roaster and getting boiling water to it would not be very logical.

Therocouple wires are two different alloy wires and if you extend them, it has to be with a like type of wire. They are also polarity sensitive so you can't reverse the polarity on them. If you do, they will do what you a re getting now for readings,
Then only success I've had with extending them without replacing the full length is using the same wire they are using, then make a good, clean mechanical connection and then use a solder with a high content of silver. Even then, the splice will you cause a little signal error but it's consistent.
 
ged
Ok... the penny finally drops! I’ve been a bit crook the last few days and I’m suffering a special kind of stupidity apparently.

I’ve got a whole array of K Type probes hanging around so I’ll load a K Type sketch and see what I come up with.

The J Type wire I’m using is very different to that installed on the Roaster. It’s using pvc insulated conductors (no doubt US made) while the Chinese wire I’m using is cloth wrapped.

I’ll give that a go and see where I get.

Thanks.
 
ged
I forgot Photobucket doesn’t work anymore...

Hopefully this will.


image.ibb.co/h0Xmm6/BB53031_A_2531_4382_A1_A5_5_BD4995_A76_EE.jpg

image.ibb.co/c4DWKR/22_D8_A6_DC_0_E75_42_FA_870_F_B60_DE5_BFF6_AD.jpg


image.ibb.co/bYfTzR/E5_E8_A5_D3_3592_4_E61_B16_D_24548_E37_C0_A4.jpg


image.ibb.co/naXmm6/FB51_BCFB_0096_4356_81_DF_019779_B587_D1.jpg
 
Will2
He probably manufactured and connected 3 thermocouples with 3 hot and 1 cold junctions.
Viliam
 
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Help in data logging and (perhaps) automating the roasting in my rottiserie oven Dataloggers/Controllers/Rate of Rise Meters 8 12/18/2019 12:58 PM
Thermocouple Type for TC4/Arduino??? Dataloggers/Controllers/Rate of Rise Meters 3 03/10/2019 8:33 AM
Using TC4C with Artisan Dataloggers/Controllers/Rate of Rise Meters 18 02/14/2019 5:39 PM
Android / IOS Data Logging App Ideas Dataloggers/Controllers/Rate of Rise Meters 13 07/30/2018 3:14 PM
Looking for TC4C board Dataloggers/Controllers/Rate of Rise Meters 124 07/17/2018 10:32 AM
Homeroasters Association Logo, and all Content, Images, and Icons © 2005-2016 Homeroasters Association - Logos are the property of their respective owners.
Powered by PHP-Fusion copyright © 2002 - 2020 by Nick Jones.
Released as free software without warranties under GNU Affero GPL v3.
Designed with by NetriX